skip to main content

Explore Research Products in the NSF-PAR

Title: Branch Prediction with Multilayer Neural Networks: The Value of Specialization
Abstract—Multi-layer neural networks show promise in im- proving branch prediction accuracy. Tarsa et al. have shown that convolutional neural networks (CNNs) can accurately predict many branches that state-of-the-art branch predictors cannot. Yet, strict latency and storage constraints make naive adoption of typical neural network architectures impractical. Thus, it is necessary to understand the unique characteristics of branch prediction to design constraint-aware neural networks. This paper studies why CNNs are so effective for two hard-to- predict branches from the SPEC benchmark suite. We identify custom prediction algorithms for these branches that are more accurate and cost-efficient than CNNs. Finally, we discuss why out-of-the-box machine learning techniques do not find optimal solutions and propose research directions aimed at solving these inefficiencies.  more » « less
Award ID(s):
2011145
NSF-PAR ID:
10249272
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
Machine Learning for Computer Architecture and Systems
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ( , 53rd Annual IEEE/ACM International Symposium on Microarchitecture)
    null (Ed.)
    Abstract—The state-of-the-art branch predictor, TAGE, re- mains inefficient at identifying correlated branches deep in a noisy global branch history. We argue this inefficiency is a fundamental limitation of runtime branch prediction and not a coincidental artifact due to the design of TAGE. To further improve branch prediction, we need to relax the constraint of runtime only training and adopt more sophisticated prediction mechanisms. To this end, Tarsa et al. proposed using convo- lutional neural networks (CNNs) that are trained at compile- time to accurately predict branches that TAGE cannot. Given enough profiling coverage, CNNs learn input-independent branch correlations that can accurately predict branches when running a program with unseen inputs. We build on their work and introduce BranchNet, a CNN with a practical on-chip inference engine tailored to the needs of branch prediction. At runtime, BranchNet predicts a few hard-to-predict branches, while TAGE- SC-L predicts the remaining branches. This hybrid approach reduces the MPKI of SPEC2017 Integer benchmarks by 7.6% (and up to 15.7%) when compared to a very large (impractical) MTAGE-SC baseline, demonstrating a fundamental advantage in the prediction capabilities of BranchNet compared to TAGE- like predictors. We also propose a practical resource-constrained variant of BranchNet that improves the MPKI by 9.6% (and up to 17.7%) compared to a 64KB TAGE-SC-L without increasing the prediction latency. 
    more » « less
  2. ; ; ; ; ( , Nature Communications)
    Abstract

    Deep neural networks (DNNs) optimized for visual tasks learn representations that align layer depth with the hierarchy of visual areas in the primate brain. One interpretation of this finding is that hierarchical representations are necessary to accurately predict brain activity in the primate visual system. To test this interpretation, we optimized DNNs to directly predict brain activity measured with fMRI in human visual areas V1-V4. We trained a single-branch DNN to predict activity in all four visual areas jointly, and a multi-branch DNN to predict each visual area independently. Although it was possible for the multi-branch DNN to learn hierarchical representations, only the single-branch DNN did so. This result shows that hierarchical representations are not necessary to accurately predict human brain activity in V1-V4, and that DNNs that encode brain-like visual representations may differ widely in their architecture, ranging from strict serial hierarchies to multiple independent branches.

     
    more » « less
  3. ; ( , Neural Computation)
    Deep convolutional neural networks (CNNs) are becoming increasingly popular models to predict neural responses in visual cortex. However, contextual effects, which are prevalent in neural processing and in perception, are not explicitly handled by current CNNs, including those used for neural prediction. In primary visual cortex, neural responses are modulated by stimuli spatially surrounding the classical receptive field in rich ways. These effects have been modeled with divisive normalization approaches, including flexible models, where spatial normalization is recruited only to the degree that responses from center and surround locations are deemed statistically dependent. We propose a flexible normalization model applied to midlevel representations of deep CNNs as a tractable way to study contextual normalization mechanisms in midlevel cortical areas. This approach captures nontrivial spatial dependencies among midlevel features in CNNs, such as those present in textures and other visual stimuli, that arise from tiling high-order features geometrically. We expect that the proposed approach can make predictions about when spatial normalization might be recruited in midlevel cortical areas. We also expect this approach to be useful as part of the CNN tool kit, therefore going beyond more restrictive fixed forms of normalization. 
    more » « less
  4. ; ; ; ; ; ; ; ; ( , Frontiers in Water)
    The water content in the soil regulates exchanges between soil and atmosphere, impacts plant livelihood, and determines the antecedent condition for several natural hazards. Accurate soil moisture estimates are key to applications such as natural hazard prediction, agriculture, and water management. We explore how to best predict soil moisture at a high resolution in the context of a changing climate. Physics-based hydrological models are promising as they provide distributed soil moisture estimates and allow prediction outside the range of prior observations. This is particularly important considering that the climate is changing, and the available historical records are often too short to capture extreme events. Unfortunately, these models are extremely computationally expensive, which makes their use challenging, especially when dealing with strong uncertainties. These characteristics make them complementary to machine learning approaches, which rely on training data quality/quantity but are typically computationally efficient. We first demonstrate the ability of Convolutional Neural Networks (CNNs) to reproduce soil moisture fields simulated by the hydrological model ParFlow-CLM. Then, we show how these two approaches can be successfully combined to predict future droughts not seen in the historical timeseries. We do this by generating additional ParFlow-CLM simulations with altered forcing mimicking future drought scenarios. Comparing the performance of CNN models trained on historical forcing and CNN models trained also on simulations with altered forcing reveals the potential of combining these two approaches. The CNN can not only reproduce the moisture response to a given forcing but also learn and predict the impact of altered forcing. Given the uncertainties in projected climate change, we can create a limited number of representative ParFlow-CLM simulations (ca. 25 min/water year on 9 CPUs for our case study), train our CNNs, and use them to efficiently (seconds/water-year on 1 CPU) predict additional water years/scenarios and improve our understanding of future drought potential. This framework allows users to explore scenarios beyond past observation and tailor the training data to their application of interest (e.g., wet conditions for flooding, dry conditions for drought, etc…). With the trained ML model they can rely on high resolution soil moisture estimates and explore the impact of uncertainties.

     
    more » « less
  5. ; ; ; ; ( , Water Resources Research)
    Abstract

    Prediction of statistical properties of the turbulent flow in large‐scale rivers is essential for river flow analysis. The large‐eddy simulation (LES) provides a powerful tool for such predictions; however, it requires a very long sampling time and demands significant computing power to calculate the turbulence statistics of riverine flows. In this study, we developed encoder‐decoder convolutional neural networks (CNNs) to predict the first‐ and second‐order turbulence statistics of the turbulent flow of large‐scale meandering rivers using instantaneous LES results. We train the CNNs using a data set obtained from LES of the flood flow in a large‐scale river with three bridge piers—a training testbed. Subsequently, we employed the trained CNNs to predict the turbulence statistics of the flood flow in two different meandering rivers and bridge pier arrangements—validation testbed rivers. The CNN predictions for the validation testbed river flow were compared with the simulation results of a separately done LES to evaluate the performance of the developed CNNs. We show that the trained CNNs can successfully produce turbulence statistics of the flood flow in the large‐scale rivers, that is, the validation testbeds.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email